Asian Journal of Agricultural Extension, Economics & Sociology

38(2): 111-119, 2020; Article no.AJAEES.55639 ISSN: 2320-7027

Analysis of the Intensity of ’s Imports from Tanzania

J. C. Ndayisaba1*, J. K. Lagat1, W. M. Ndayitwayeko2 and S. K. Kiprop3

1Department of Agricultural Economics and Agribusiness Management, Egerton University, P.O.Box 536, Egerton-20115, . 2Department of Rural Economics, University of Burundi, Burundi. 3Department of Economics, Egerton University, P.O.Box 536, Egerton-20115, Kenya.

Authors’ contributions

This work was carried out in collaboration among all authors. Author JCN designed, wrote the first draft of the manuscript and performed statistical and econometric analysis. Author WMN revised and improved the statistical and econometric analysis. Authors JKL and SKK revised the whole manuscript and improved it before its submission. All authors read and approved the final manuscript.

Article Information

DOI: 10.9734/AJAEES/2020/v38i230317 Editor(s): (1) Dr. Roxana Plesa, University of Petrosani, Romania. Reviewers: (1) Paul Kweku Tandoh, Kwame Nkrumah University of Science and Technology, Ghana. (2) Hussin Jose Hejase, Al Maaref University, Lebanon. Complete Peer review History: http://www.sdiarticle4.com/review-history/55639

Received 18 January 2020 Accepted 22 March 2020 Original Research Article Published 04 April 2020

ABSTRACT

Neighboring countries usually exchange goods and services, taking advantage of proximity and other shared socio-economic characteristics among citizens. This study explored the intensity of Burundi’s rice imports from Tanzania. First of all it determined and analyzed the evolution of the intensity of rice imports. Secondly, it estimated the factors which influence the intensity of rice imports. The results indicated that Burundi’s rice imports from Tanzania remained more intensive over 2003-2018. However the financial crisis which slapped the world in 2007-2008 harmed Burundi’s rice imports from Tanzania probably due to lack of finance to import. Moreover, the results revealed that the Burundian national income, its exchange rate and its trade openness significantly had a positive effect on the intensity of Burundi’s rice imports from Tanzania.

Keywords: Imports; neighboring countries; intensity of trade; food trade. ______

*Corresponding author: E-mail: [email protected];

Ndayisaba et al.; AJAEES, 38(2): 111-119, 2020; Article no.AJAEES.55639

1. INTRODUCTION neighboring countries located in the eastern part of Africa. Burundi is a Around the world, food trade considerably bordering with in North, Democratic contributes to the improvement of citizens’ Republic of the Congo in West and Tanzania in welfare. Through trade, food commodities are both East and South sides. Burundi was self- exchanged from countries with surplus to sufficient in food production before 1990’s [5]. countries challenged with food shortage. This Nevertheless, the civil war which struck the results in enhanced food supply leading to country over the period 1993-2005 reduced the sustainable food access in countries involved in output from all sectors, included. trade. Salvatore [1] indicates that we live a Since then, Burundi relies on imports to offset its globalized world where tastes converge and deficit in domestic food needs. As an illustration, goods and services we usually use are provided Table 1 provides a view of the balance between by foreigners. Therefore, food trade’s existence Burundi’s exports and imports in selected food matters more in order to ease a smooth access commodities suchlike cereals, meats products to food commodities and hence improve the and dairy products for the period 2010-2015. welfare of citizens. From Table 1, it is evident that Burundi faces a Empirical studies pointed out that international deficit in the mainly consumed food commodities. trade benefits all countries and this leads to a In such a situation, imports step in to offset that sustainable development across nations. existing deficit. In Burundi, food commodities are Marshall [2] indicates that the causes which imported from neighboring countries, from the determine the economic progress of nations region and from the rest of the world mostly Asia belong to the study of international trade. and/or Europe. This study was meant to explore Examining the existing nexus between trade and Burundi’s rice imports from Tanzania using an , Robertson [3] described intensity of trade approach. More specifically, this exports as an engine of growth and Minford et al. study firstly determined the intensity of Burundi’s [4] hailed foreign trade as an elixir of economic rice imports from Tanzania. Thereafter, it growth.Therefore, no matter how far countries estimated the factors which influence the are located from each other, still exchange of intensity of rice imports. The broad aim of this goods and services takes place and benefits to study was to contribute towards improved all countries involved in trade. Trade existence performances of Burundi’s rice imports from among nations is triggered by the phenomenon Tanzania. This study covered the period 2003- of globalization which intensifies day after day 2018 due to data availability. particularly in open economies. Tanzania was chosen over other neighboring Countries facilitate trade by partially or definitely countries with Burundi reason being that it removing tariffs and other non-tariffs barriers shares a large physical border with Burundi (from which are perceived to significantly hamper the East to South), with many entry points along the movement of goods and services between them. shared physical border. This is perceived as a Likewise, neighboring countries usually trade comparative advantage as far as imports and/or each other and establish some forms of exports are concerned. Rice was chosen among agreements easing trade between them. It is in other food commodities due to the existing high this context that this study sought to analyze rate of rice consumption in Burundi caused by a trade in rice between Burundi and Tanzania, two sporadic increase in number and extension of

Table 1. Burundi’s total imports and exports (in 1000$) for cereals, meat products and dairy products and eggs in 2010-2015

Year Imports Exports Balance Cereals Dairy Meat Cereals Dairy Meat products and products products and products eggs eggs 2010 27031.222 1850.546 74.238 829.249 4.596 0.000 Deficit 2011 47764.465 4133.476 46.174 392.610 0.000 0.000 Deficit 2012 100870.933 2651.12 56.528 5.285 0.989 0.000 Deficit 2013 47230.166 1937.866 103.596 2915.713 25.125 3.317 Deficit 2014 42276.060 1905.181 4811.558 4807.307 0.675 2.704 Deficit 2015 28756.55 1249.992 3105.931 5636.861 0.172 0.006 Deficit Source: UNCOMTRADE, 2019

112

Ndayisaba et al.; AJAEES, 38(2): 111-119, 2020; Article no.AJAEES.55639

urban centres. In Burundi, there are few theory indicates that trade takes place between empirical studies related to food trade which countries with the same level of development. have been carried out so far. Empirical studies The similarity can be seen in the aspects of done are more general and did not use the location, culture, political and economic interests intensity of trade approach, to analyse trade and natural resources among other aspects. The flows between Burundi and Tanzania as far as relevance of this theory in the context of this rice imports are concerned. Thus, this study was study lies in the fact that the countries involved in meant to bridge that gap left by previous this study (Burundi and Tanzania) share many empirical studies. The results of this study firstly features and this makes them to be similar in a provided relevant quantified information to policy way or another. makers in Burundi and Tanzania. Such information could be more useful while Some empirical studies conducted on the drivers addressing an issue related to rice demand and of the intensity of trade are oriented at firm-level. supply across the two neighboring countries. As an illustration, we can mention Schlegelmilch Secondly, the results feed into existing literature and Crook [9] who estimated the determinants of on bilateral trade between neighboring countries. export intensity at firm level. Their study employed a multiple regression analysis and 1.1 Overview of Production, Consump- found out that the export intensity is negatively tion and Imports of Rice in Burundi related to the domestic growth of sales. Seyoum [10] explored the determinants of import intensity There are three ecological zones in Burundi in US foreign trade zones. Factors affecting where rice is grown: the irrigated Imbo plain, the import intensity were categorized into three rain fed (non-irrigated) areas of Imbo and Moso groups suchlike external factors, firm lowlands and the non-irrigated areas of the characteristics and firm business strategy. elevated marshland region [6]. Concerning Findings reveal that the most promising rice consumption in Burundi, the rate followed predictors of import intensity of firms operating in an upward trend in these years. Rice is USA free trade zones are the policy environment consumed by households and mostly by in the form of inverted tariff benefits and firm communities suchlike military camps, schools, business strategy. Other empirical studies prisons and religious communities. In oriented at country level were carried out so far. households, preferences depend on household This is the case of Elmorsy [11] who analysed income. Nzeyimana [7] indicates that the the determinants of trade intensity of Egypt with demand for rice has increased for about 70 % in COMESA countries. Both the intensity of trade 2013 and registered a peak in 2010. and the gravity model was used. Findings show that the was the main Although there is a sporadic increase of rice factors which affect the intensity of trade demand in the country, domestic rice production between Egypt and COMESA countries. fails to meet local demand. In such a situation, the country relies on imports. Rice is imported 3. MATERIALS AND METHODS from the world’s top rice producers such as , , Thailand and Vietnam for about 78% The methods used to meet the objectives of this and the rest is mostly imported from the EAC study are grouped into two section. The first region. However, Tanzanian rice is the most section deals with methods used to determine frequently found in Burundi’s local markets the intensity of imports and the second section is although this does not appear in official statistics about methods utilized to estimate the [7], the proximity with Tanzania being one of the determinants of intensity of rice imports. reasons. Most of households in Burundi appreciate the taste of Tanzanian rice. 3.1 Determination of the Intensity of Burundi’s Rice Imports 2. THEORETICAL FRAMEWORK In this study, the intensity of Burundi’s rice This study was grounded by the theory of country imports was determined. This is a descriptive similarity. It was developed by Steffan Linder in approach helping to deeply understand how 1961 [8]. The theory stresses that a nation intensively Burundi imported rice from Tanzania. exports those manufactured products for which a The trade intensity technique was developed by large domestic market exists. In other words, the Kojima in 1964 [12]. The intensity of trade is

113

Ndayisaba et al.; AJAEES, 38(2): 111-119, 2020; Article no.AJAEES.55639

defined as a ratio of two export shares [13]. The linear regression containing these factors was numerator is the share of the destination of estimated. Hence, the expression of the equation interest in the total exports. The denominator is of factors affecting the intensity of Burundi’s rice the share of the destination of interest in the imports from Tanzania was given by the following exports of the world as a whole. Ambrose and equation. Sundar [12] argued that the intensity of trade does not suffer from any size bias and one can LnMI k     X   compare the statistic across regions and over ijt 0 it it t (2) time. Across countries, there are empirical studies on intensity of trade which have been The factors associated with the intensity of carried out so far. To determine the intensity of Burundi’s rice imports were analysed using two trade, this study adopted [12] for a reason that it quantitative techniques suchlike Unit root test is the recently improved formula to determine and Johansen Maximum Likelihood test of trade intensity. Therefore the expression of the Cointegration. The latter is very crucial since the intensity of Burundi’s imports from import presence of a unit root in the series is likely to Tanzania was given by the following expression: lead to a spurious regression as the structure of the error term is unknown and the possibility of unbiased estimates exists. The specification of the model was done in searching the = ∗ 100 (1) determinants that influence the intensity of imports and the model chosen was as follows:

where: k ln MIijt  0  1it log gdpbuit   2it logexratiit  i and j respectively refer to Burundi and (3) 3it logbuagrilandit   4it logtradeopen4it   t Tanzania; k refers to rice; t refers to time. The loggdpbuit is the variable of GDP of Burundi denotes intensity of imports between i at the period t predicted to have a positive and j at time t. influence on the intensity of Burundi’s rice imports; logexratiit is the exchange rate of the (BIF) in terms of USA dollar denotes i’s imports (in values) from j for commodity k at time t. currency which was predicted to have a neutral influence depending on the depreciation or appreciation of the BIF against the USA denotes i’s total imports (in values) for commodity k at time t. dollar currency; logbuagriland is the arable land allocated to agricultural production in overall. denotes total exports (in values) to Tanzania at time t. The extension of arable land of Burundi was predicted to have a negative effect on the denotes Tanzania total export (in intensity of Burundi’s rice imports. Given that values) at time t. there is an increasing trend of highland rice production, this was predicted to denotes total i’s export (in values) at time t. eventually curb the intensity of rice imports if the domestic rice production increases In the above equation is subtracted from tremendously. Logtradeopen is a variable for a reason that a country cannot export denoting trade openness of Burundi. The trade goods and services to itself. The only share policy instruments used to control trade are tariffs it can meaningfully have in total world trade and quota and a battery of non-tariff barriers. is a share in the imports of all other countries Overall, tariff is commonly used by Burundian [14]. trade policy makers but the limit of such instrument is within the East Africa Community’s 3.2 Factors Affecting the Intensity of agreed tariff limit on raw materials, semi-finished Burundi’s Rice Imports and finished products. This variable was predicted to have a positive effect on the In this study, the determinants of the intensity of intensity of Burundi’s rice imports. We choose Burundi’s rice imports were estimated. A multiple the Johansen co-integration procedure [15]. This

114

Ndayisaba et al.; AJAEES, 38(2): 111-119, 2020; Article no.AJAEES.55639

procedure follows a Maximum Likelihood Trade Solution (WITS), data on GDP and arable Approach (ML) and is formulated as follows: agricultural land were found from database, data on exchange rate was retrieved Z   Z  ...  Z  Z  X  u (4) t 1 t1 k1 tk 1 tk t t from Burundi Central Bank (BRB) and other databases where data was retrieved from include Where Zt is a vector of I (1) endogenous the Burundian bureau of statistics (ISTEEBU) variables, ∆Z=Zt-Zt-1 and Xt is a vector of I (0) and Food and Agriculture Organization database exogenous variables. Γi, Π and Ψ are (n x n) (FAOSTAT). The data were transformed into vector matrices of parameters. The above logarithmic function due to the model used for equation (4) contains the short-run and long-run analysis. adjustment to changes in Zt represented by Γi and Π respectively. In co-integration analysis, the 4. RESULTS AND DISCUSSION number of co-integrating relationship are given by the rank of Π and are denoted by r and trace The results were discussed into two sections. statistics are used to test the null hypothesis of The first section addresses the first objective of at most r co-integrating vectors against the this study which was to explore the evolution of alternative that the number of co-integrating the intensity of Burundi’s rice imports from vectors is greater than r. Another step followed in Tanzania. The second section addresses the this study was to estimate the Vector Error second objective of this study which was to Correction Model (VECM) in order to examine estimate the determinants of the intensity of the existence of the long-run and short-run Burundi’s rice imports. relationship among variables through autoregressive lags following [16]. The Vector 4.1 Results of Intensity of Burundi’s Rice Autoregressive (VAR) had to be turned into Imports from Tanzania equation (5) of VECM. The results on the intensity of Burundi’s rice Z     Z   Z  ...  Z   (5) imports from Tanzania are presented in Table 2. t 0 1 t1 2 t2 k tk t From Table 2 it is evident that Burundi’s rice And imports from Tanzania were more intensive except for the period ranging between 2008 and Z   Z   Z ...  Z  Z   (6) t 1 t1 2 t2 k1 t(k1) tk t 2009 where low level of intensity was observed. Possible explanations to this could be the 3.3 Data Sources financial crisis of 2008-2009 in the World. Examining the evolution of the intensity of rice This study used time series data. Data on rice imports over time, we used histograms to depict imports was collected from World Integrated the evolution. The results are presented in Fig. 1.

Table 2. The intensity of Burundi’s rice imports from Tanzania

Year Country Tradingpartner Commodity Intensity of imports (%) 2003 Burundi Tanzania Rice 362.4120059 2004 Burundi Tanzania Rice 312.1094954 2005 Burundi Tanzania Rice 398.4977985 2006 Burundi Tanzania Rice 262.7569611 2007 Burundi Tanzania Rice 182.3257096 2008 Burundi Tanzania Rice 48.29880241 2009 Burundi Tanzania Rice 25.82167978 2010 Burundi Tanzania Rice 114.462341 2011 Burundi Tanzania Rice 311.2005163 2012 Burundi Tanzania Rice 143.9186425 2013 Burundi Tanzania Rice 364.5050608 2014 Burundi Tanzania Rice 403.1724799 2015 Burundi Tanzania Rice 380.0051073 2016 Burundi Tanzania Rice 401.3768169 2017 Burundi Tanzania Rice 369.8789714 2018 Burundi Tanzania Rice 326.0534814 Source: Authors

115

Ndayisaba et al.; AJAEES, 38(2): 111-119, 2020; Article no.AJAEES.55639

450 400 350 300 250 200 bdi vs tz 150 100 50 0

Fig. 1. Evolution of the intensity of Burundi’s rice imports from Tanzania in 2003-2018 Legend: bdi: Burundi; tz: Tanzania

Table 3. Unit root test

Unit root test Level lnMI lngdpbu Lnexrati lnbuagriland Lntradeopen ADF Test Level -3.230 -3.940 -2.646 -1.573 -2.169 (0.0183)* (0.0018)* (0.082) (0.4973) (0,2176) 1st Difference - - -2.502 -1.569 -2.199 (0.1150) (0.4993) (0.2066) Philips-Peron Level -3.246 -3.954 -2.188 -1.705 -2.182 Test (0,0174)* (0.0017)* (0.2105) (0.4286) (0.2130) 1st Difference -2.922 -1.638 -2.228 (0.1745) (0.4632) (0.1962) Note: (*) [17] one-sided p-values for the reject hypothesis of unit root at 5 percent

From Fig. 1, it is clear that the intensity of method enables us to know the number of co- Burundi’s rice imports followed an upward trend. integration relationships that remain between our However, an exception was observed around the long-term variables. The maximum likelihood period between 2008 and 2010 whereby the method was used and the results obtained are trends followed a decreasing slope for the presented in Table 4. reasons explained in previous paragraphs. Results from Table 4 reveal the existence of 4.2 Results of Factors Affecting the long-run relationship among the variables. If the Intensity of Burundi’s Rice Imports trace statistics and the maximum Eigen statistics from Tanzania are greater than the 5% critical values, the null hypothesis of no-cointegration is rejected in favour of the alternative hypothesis at the level. Before embarking on the estimation of factors The trace statistic and the Maximum Eigen which affect the intensity of rice imports, tests statistic show that there is co-integration among related to the nature of data used were the variables implying a long run equilibrium performed. Dealing with time series data relationship. Table 4 shows that at least there are requires to test for unit root so that we know the 4 co-integration equations. Hence, there is a long stationary and non-stationary variables. The run equilibrium relationship between the intensity results obtained using STATA 15 are presented of imports and related independent variables in in Table 3. the rice sub-sector of Burundi. Table 3 shows that for three variables, the statistics of the ADF and PP tests are lower than 4.3 Summary Statistics (1993-2018) the criterion statistics of the different thresholds than after a prior differentiation, so they are The intensity of rice imports is far above zero and integrated with orders I (1) and we conclude that reveals high trade intensity between Burundi and there may be a co-integrated relation. We then Tanzania. Since the bilateral negotiation exists carried out a co-integration Johanson test. The between the two countries, the leading theme in

116

Ndayisaba et al.; AJAEES, 38(2): 111-119, 2020; Article no.AJAEES.55639

negotiation was to find out ways on how to Burundi. Though land-locked, Burundi has taken revamp the cooperation and development of an opportunity offered by the EAC regional trade trade along the central corridor. Table 5 provides integration by open up to other countries for summary statistics of the intensity of trade. trade as evidenced by the indicator of trade openness. In order to estimate the determinants Before Burundi used to have access to Indian of the intensity of rice imports, we carried out a Ocean through the port of Mombasa but recently vector error correction model to find out short-run Dar-Es-Salaam becomes the right choice for and long-run effects when equilibrium holds. The export and import goods from abroad. This results of dynamic VEC model present rich and strategy of diversifying trade had to help Burundi interesting findings. These results are presented to solve its salient challenge of land-locking in Table 6. There is long-run causality running state. The statistics show obviously that the from gdpbu to trade intensity. economic size of Tanzania is four time that of

Table 4. Johanson maximum likelihood test of co-integration, unrestricted co-integration rank test (Trace and maximum Eigen values)

Level Trace Maximum Eigen value Hypothesized no. of CEs Statistics 5% critical value Statistics 5% critical value None 181.80 94.15 80.60 39.37 At most 1 101.30 68.52 45.13 33.46 At most 2 56.17 47.21 25.15 27.07 At most 3 31.01 29.68 21.56 20.97 At most 4 9.46* 15.41 9.29 14.07 At most 5 0.17 3.76 0.17 3.76 Number of observation: 24 Lags: 2 Notes: Trace and Max-Eigen value test indicate 1 cointegrating eqn(s) at the 0.05 level * denotes rejection of the hypothesis at the 0.05 level [18] p-values

Table 5. Summary statistics

Variable Obs Mean Std. Dev. Min Max Intensity 26 546.1839 981.617 19.71346 3717.735 gdpbu 26 23.19585 7.068962 1.000186 33.16152 gdptz 26 82.96239 31.49074 37.68172 140.3346 exrate 26 19.4653 4.855291 10.86137 25.73841 buagrland 26 74.89965 4.188339 67.48443 82.67135 tzagrland 26 40.78571 2.930922 36.80289 44.81824 tradeopen 26 3.467752 .2758828 3.042809 3.854394 Note: gdpbu and gdptz (in million USA$) Where: gdpbu= GDP of Burundi, gdptz= GDP of Tanzania, exrate= Exchange rate of Burundi (1US$ to Burundi francs), buagriland= Agricultural land or arable land of Burundi, tzagrland= Agricultural land of Tanzania and tradeopen= Trade Openness

Table 6. Results of Vector Error Correction Model (VECM)

Short-run relationship Coefficient Z P-Value Logintensity 1.00

Loggdpbu 9.10 13.63 0.00

Exrati 20.88 4.67 0.00

Logbuagriland 56.28 8.13 0.00

Tradeopen 8.97 15.23 0.00 ECM -0.72 -2.58 0.01 Log likelihood = 1641.37, Det (Sigma_ml) = -1.08e7

117

Ndayisaba et al.; AJAEES, 38(2): 111-119, 2020; Article no.AJAEES.55639

The model sought to relate the trade intensity 5 . C ONCLUSION AND POLICY IMPLICA- in rice to its explanatory variables such as T I O NS GDP of Burundi (gdbbu), exchange rate (exrati), land allocated to agriculture in T h e r e sults from this study revealed that the Burundi (buagriland) and indicator of openness intensity of Burundi’s rice imports from Tanzania (tradeopen). The results revealed that there w a s t o o high. Hence, we conclude saying that is a long-run causality running from the B u r u n d i intensively imports rice from Tanzania listed explanatory variables to the intensity of and the trend of the intensity of rice imports import. follows an upward slope. However, an exception was observed in the period between 2007 and The adjustment error correction of -72% is 2010 whereby the intensity of Burundi rice significant and indicates a higher speed imports decreased. This decrease could be adjustment (that is, the speed at which the attributed to the financial crisis erupted in the deviation from long-run equilibrium is adjusted world causing a lack of finance to import. slowly where 72% of the disequilibrium is Furthermore the results pointed out that the removed each period). This shows that, the economic performance boosts the intensity to speed of adjustment to where rice imports trade in rice. Arable land expansion, trade equilibrate the real Gross Domestic Product in openness and appreciation of exchange rate Burundi is at the rate of 72%. In the short run, h a v e positive effects on the trade intensity all explanatory variables are positive and between Tanzania and Burundi. We here statistically significant (p<0.01). For instance, a forecast that the findings are the precursor of the unit increase in GDP of Burundi, land allocated intensity of trade in other sectors of economy if to , exchange rate and the central corridor will be developed and the trade openness leads to about more than finalisation of of railway linking proportionate increase in the intensity of Tanzania, Burundi and Democratic Republic of Burundi’s rice imports from Tanzania. When the the Congo will castigated a more dynamic Burundian franc appreciates, this causes an and prosperous regional trade in Great Lakes expansion of Burundi’s rice imports from its region. neighbour (Tanzania). Albeit the political instability of 2015, the I n li n e with these results, trade policy makers (a member of East African Community) has should then encourage trade between the two slightly increased since 2014 and settled at n e ig h b ouring countries. They should establish an 3.8 percent in 2018. The economic performance efficient policy framework favouring trade by is attributed to the performance of primary r e m o ving all impediments to trade between the sector that is agriculture, the increasing two countries. However, provided that Burundi production of mining sector and diversification of agronomic conditions are favourable to rice, exports. Since Burundi has put in place trade policy makers in the agricultural sector should policies that favour trade flows, the take appropriate actions to optimally boost the implementation of such policies has promoted level of domestic rice production, to prevent the both the intensity of trade and economic growth country from being more depending on rice on overall [19]. imports. The money used in importing rice to offset the domestic production could be used in These results go along with those of [20] which other sectors in order to enormously contribute to states that the key drivers of production and the economic growth of the country. Burundi demand, including trade and related policies, should manage well the exchange rate policies shape the patterns of trade with potentially since this study shows a close relationship important implications on food security. The trade between exchange rate and the intensity of rice link between Tanzania and Burundi is due to the imports. greater participation of Burundi in regional trade (EAC trade). This enshrined in Burundi’s national ACKNOWLEDGEMENT trade strategies and good management of the process of trade openness and at the end, trade Authors acknowledge the Centre of Excellence in obviously works in favour of economic growth Sustainable Agriculture and Agribusiness and food security as evidenced by the results of Management of Egerton University for its this study. support.

118

Ndayisaba et al.; AJAEES, 38(2): 111-119, 2020; Article no.AJAEES.55639

COMPETING INTERESTS 10. Seyoum B. US foreign trade zones and import intensity examining determinants of Authors have declared that no competing import intensity in US foreign trade zones. interests exist. European Business Review; 2017. 11. Elmorsy SS. Determinants of trade REFERENCES intensity of Egypt with COMESA countries. Bandung. 2015;2(1):1-25. 1. Salvatore D. International economics: 12. Ambrose B, Sundar RP. A brief analysis of Trade and finance. 11th Edition. Fordham India-Japan bilateral trade: A trade University, John Wiley and Sons Inc; 2014. intensity approach. International Journal of 2. Marshall A. Principles of Economics. 8th Economics, Commerce and Management. edition. Macmillan for the Royal Economic 2014;4(2):1-7. Society; 1959. 13. Mikic M, Gilbert J. Trade statistics in policy 3. Robertson DH. The future of international making-A handbook of commonly used trade. Economic Journal. 1938;48(189). trade indices and indicators. United 4. Minford P, Riley J, Nowell E. The elixir of Nations ESCAP, New York; 2008. growth: Trade, non-traded goods and 14. Peter D, Ross G. Trade intensities and the development. Centre for Economic Policy analysis of bilateral trade flows in a many Research Discussion Paper No. 1165; country world: A survey. Hitotsubashi 1995. Journal of Economics. 1982;22(2):62-84. 5. Ndayitwayeko WM, Mark OO, Mark K, 15. Johansen S. Statistical analysis of Philip ON. Exploring agri-food import cointegration vectors. Journal of Economic dependency of Burundi: A gravity model Dynamics and Control. 1988;12(2-3): approach. Journal of Economics and 231-254. Sustainable Development. 2014;5(14): 16. Hallam D, Zanoli R. Error correction 90-100. models and agricultural supply 6. International Rice Research Institute response. European Review of Agricultural (IRRI). Rice production in Burundi. Economics. 1993;20(2):151-166. Unpublished document; 2009. 17. MacKinnon JG. Numerical distribution 7. Nzeyimana M. Market integration and price functions for unit root and cointegration transmission of rice in Burundi. Master of tests. Journal of Applied Econometrics. Science dissertation, Moi University; 2016. 1996;11:601-618. 8. Neculita M, Sarpe DA. International 18. MacKinnon JG, Haug AA, Michelis L. commerce-development factor in Numerical distribution functions of and India. Annals of the University likelihood ratio tests for cointegration. Dunarea de Jos of Galati: Fascicle: I, Journal of Applied Econometrics. 1999;14: Economics and Applied Informatics. 2018; 563-577. 24(3):65-72. 19. Central Bank of Burundi (BRB). Annual 9. Schlegelmilch BB, Crook JN. Firm‐ report. , Burundi; 2019. level determinants of export intensity. 20. Food and Agriculture organization. Managerial and Decision Economics. Retrieved from FAO Statistical database 1988;9(4):291-300. on 26 February; 2020. ______© 2020 Ndayisaba et al.; This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Peer-review history: The peer review history for this paper can be accessed here: http://www.sdiarticle4.com/review-history/55639

119